In Liberia we could distinguish
about three general strains of Basenjis. Within these strains there was
still more variation, and our group of interested owners were
constantly involved in a selective breeding programs. At the same time,
we all attempted to study the genetics of various traits. This was
extremely difficult because we were continually adding native lines in
which we had to rely entirely on phenotype.
only studied character which appeared to be simple and not quantitative
was coat color. We felt that it was controlled by a single pair of
genes with black dominant to red. In our Liberian breedings, a black x
red cross usually gave a 3:1 ratio of black to red. (To my knowledge no
tri-colors were ever produced and we did not see or hear of any in the
country.) This would be a classic Mendelian ratio of single gene
complete dominance. After returning to the United States and discussing
the problem with Geneticists, I began to doubt our concept of complete
dominance; and the recent red Fula
line crosses which produced blacks supported the doubts. Modern
Geneticists are concluding that very few characters follow the simple
Mendelian ratios, so I would like to suggest the following possible
means of coat color inheritance.
It seems probable
that the Basenji coat color must be produced by a multi-gene complex
with black lacking complete dominance (i.e., having incomplete
dominance). As an arbitrary case, suppose that the color is due to
three gene pairs which we will label A, B,
and C. Further suppose that to be black a Basenji
must carry one dominant in each gene pair. The other component of the
pair could be either dominant or recessive without affecting the color.
Thus Aa Bb Cc, AA BB CC, Aa
BB Cc and a few other combinations would all give black
phenotypes; all that is needed is one dominant in each gene pair.
In this type of inheritance pattern, then, all other gene
combinations would be red: Aa Bb cc, AA
bb CC, aa bb cc, etc. As long as one gene
pair lack a dominant, the result is a red coat color.
Then theoretically, a red AA Bb cc
crossed with a black AA BB CC would yield black
progeny in a 3:1 ratio of black to red. This would explain our
observations in Liberia which we interpreted as simple complete
dominance. The two matings of red American Basenjis into our Kiki line have followed this
Using the same scheme, as red Aa
Bb cc crossed with a red aa bb Cc could
produce black offspring (genotype Aa Bb Cc). The
ratios from such a cross would 7 red to 1 black. Possibly this is how
the blacks were produced in the Fula line crosses.
Again using the same scheme, it would be possible to cross a red AA
BB cc and a red Aa Bb cc and get no
blacks at all. Perhaps this explains why no blacks occurred before the
Kiki and Fula lines were introduced.
All the above
postulations are probably over simplified. For instance, there may be 4
or more gene pairs involved instead of 3 and this would considerably
reduce all of the above frequencies. These schemes are offered for
comments and criticisms by persons with more experience in Basenji
In Liberia we also attempted to study
the genetics of the bark, but learned little except that it seems
quantitative. We avoided dogs which had an "appreciable bark," but
since no one knew where the line was drawn between a "woof" and a
"bark", decisions were difficult. I now realize that our standards on
this point were not as critical as those of Basenji breeders here.
Our selections for bone structure and general conformation
were rather good, partly because these characters were easily
recognized. All of the Liberian lines are long in the loin, but not
having other lines for comparison, most of us did not realize this.
The curl of the tail was always a problem; the tails in the
Liberian strains seldom curled as tightly as we would liked. As I
mentioned before, the tail is of little importance to the native who
sometimes docks it altogether. An interesting commentary on this fact
can be seen on a Liberian postage stamp issued for the 1960 Olympic
Games. Pictured on the stamp is a typical native hunter with his
typical country dog - a beautiful Basenji, but a nearly straight tail.
In comparing the Liberian Basenjis with the Basenji in
America, there is a noticeable difference in temperament. The Liberian
dog seems more relaxed, docile and easily trained. Oddly enough, this
may be related to a legal characteristic of the tribal Liberian. Most
natives enjoy legal "palaver" and will sue their neighbor at the
slightest provocation. A dog which chases the neighbors chickens or
snaps at the children can cost the owner several months wages. So here
again the native system of selection comes into play - they eat the
dogs that won't mind. Perhaps the true reason for this difference in
temperament is more complex. Lawsuits, as a common practise, seem to
have reached the tribal areas only in the last generation; although
they may have been developing along with the nationalistic movement for
about 60 years. This may not have been enough time to develop such a
trait in the dogs; it may only have been factor in maintaining it.
I should point out that none of these remarks is strongly
significant genetically. They represent observations of possibly 50 or
60 specimens at the most and seldom from repeated matings. To establish
validity, we should have a minimum of 100 individuals from the same
mating. Of course this is a virtual impossibility with Basenjis,
requiring about 20 repeated matings, but the validity could be
increased by observations from 3 or 4 repeated matings.